Currently prognostic and predictive cancer markers are inadequate, resulting in over or under treatment of a large percentage of patients, particularly those at early stages of diseases. The goal of this proposal is to evaluate the perinucleolar compartment (PNC), a multi-component cellular structure, as a prognostic and predictive tumor marker in a pre-clinical setting. Our preliminary findings demonstrate that PNC prevalence (the percentage of cells contains at least one PNC) associates with the malignant phenotype both in cultured cells and in vivo in breast tumors. PNC prevalence increases in parallel with the disease progression of breast cancer and is highest in distant metastases. Multivariate analyses (adding onto size and grade) demonstrate that breast cancer patients having primary tumors with a PNC prevalence below 23% show a significantly better disease free survival rate than those above 23% in the cases with negative nodes or <4 positive nodes. Furthermore, the PNC is highly sensitive to a subset of chemotherapeutic drugs. These findings suggest that the PNC reflects the overall malignant behaviors of cancer cells and contains independent prognostic value, thus it could be an ideal candidate as a tumor marker. Work proposed in this grant will test this hypothesis. The Specific Aims will evaluate and confirm the prognostic power of PNC prevalence by examining the relationship between this marker and disease free and overall survival rate in similarly diagnosed breast cancer patients. Studies will also assess the predictive power of PNC prevalence by comparing the outcomes of patients who are treated with drugs that reduce PNC prevalence in cultured cells vs. the outcomes of those untreated or treated with drugs that have no effect on PNCs. We expect the proposed studies will clarify whether the PNC, an easily detectable cellular structure, could improve the accuracy of breast cancer prognosis and prediction when used in combination with existing markers.